JP2019086349A - Condition monitoring device and abnormality diagnostic method for bearing - Google Patents

Abstract

To provide an abnormality diagnostic method and a condition monitoring device that have achieved an improvement in accuracy required for detecting abnormality of a rotary machine under a noisy environment.SOLUTION: An abnormality diagnostic method comprises the steps of: setting a main monitoring band on each of a fundamental frequency and higher harmonic frequencies that, resulting from a damage of a bearing, are calculated based on a rotation speed signal and a design specification of the bearing (S3); and applying a frequency analysis to measurement data, and determining abnormality of the bearing to be a diagnosis object using a first value Σm that indicates a total of powers of frequency components in the main monitoring bands and a second value Σo that indicates a total of powers of frequency components in sub-monitoring bands other than the main monitoring bands, at each of the fundamental frequency and the higher harmonic frequencies (S5 to S9).SELECTED DRAWING: Figure 2

Description

  The present invention relates to a rolling bearing state monitoring device and an abnormality diagnosis method.

  Conventionally, in an object to be tested such as a rotating machine, the object to be tested is measured based on measurement data of physical quantities such as acceleration, velocity, displacement, sound, AE (Acoustic Emission), and electric power measured by various vibration sensors. I am monitoring the status. Specifically, a feature amount indicating the feature of the waveform indicated by the measurement data is calculated, and the presence or absence of an abnormality of the test object is determined using the calculated feature amount.

  A vibration system sensor is attached for the purpose of diagnosing abnormality of the rotating parts, and it is based on the rotational speed signal and the frequency component due to the damage of the rotating parts calculated based on the design specifications and the signal waveform detected by the vibration system sensor There is known a method of comparing and collating with frequency components of measured data based on the above, and identifying the presence or absence of abnormality of the rotating part and the abnormal part and part based on the comparison result (refer to JP2006-077945A).

JP, 2006-077945, A Unexamined-Japanese-Patent No. 2001-021453

  However, with such conventional methods, although it can be determined to some extent when measuring vibration in a low noise environment (such as a laboratory), a noisy environment (where there is a large power electrical facility around the object to be measured, When the vibration is measured in a place where there is impact vibration etc., the determination is often difficult due to the inclusion of noise unrelated to the abnormality. If a misdiagnosis occurs, it leads to a reduction in the reliability of the abnormality diagnosis.

  The rotary machine is an important equipment most commonly used in the industry, and its troubles and breakdowns have a great adverse effect on production and quality. Although vibration diagnosis is the mainstream for abnormality diagnosis of rotating machines because of its simplicity, it is often installed in an environment with a lot of noise, so it is important to take measures against noise.

  In addition to the above-mentioned abnormal frequency band comparison and comparison method, a determination method using an OA value (Overall Value) is also known (see Japanese Patent Application Laid-Open No. 2001-021453). The OA value means fast Fourier transform (FFT) of a time signal such as vibration to obtain a power spectrum, and means the combined power of all analysis frequencies (product sum of power spectrum of each analyzed frequency). However, in the case of this method as well, the determination becomes difficult due to the inclusion of noise.

  The present invention has been made to solve the above-described problems, and an object thereof is to provide an abnormality diagnosis method and a state monitoring apparatus with improved accuracy in detecting an abnormality of a rotating machine in a noisy environment. It is to be.

  The present invention, in summary, is an abnormality diagnosis method for determining the presence or absence of an abnormality in a bearing based on measurement data obtained by measuring mechanical vibration generated at the time of rotation of the bearing using a vibration system sensor. The step of setting the main monitoring band to each of the fundamental frequency and the harmonic frequency of the fundamental frequency caused by the damage of the bearing calculated based on the original, and performing frequency analysis on the measured data, the fundamental frequency and the harmonic frequency The diagnosis is performed using a first value indicating the sum of powers of frequency components in the main monitoring band and a second value indicating the sum of powers of frequency components in the sub-monitoring bands other than the main monitoring band in each of Determining an abnormality of a target bearing.

  Preferably, the determining step determines that the bearing to be diagnosed is abnormal if a value obtained by dividing the first value by the second value exceeds a threshold.

  Preferably, the vibration system sensor is at least one of an acceleration sensor, an acoustic sensor, an ultrasonic sensor, and an AE sensor.

  The present invention, in another aspect, is a state monitoring device that performs abnormality diagnosis using any of the above-described abnormality diagnosis methods.

  Preferably, the state monitoring device includes a filter unit that removes a signal of a frequency band not to be analyzed from the signal waveform detected by the vibration sensor.

  In the present invention, identification of a bearing abnormality is facilitated by utilizing the proposed special analysis parameter, particularly for bearing diagnosis when low frequency noise is mixed, which is generally considered to be difficult. Thereby, abnormality diagnosis of a bearing etc. can be implemented with high reliability automatically.

It is a block diagram which shows the structure of the state monitoring apparatus which concerns on this Embodiment. It is a flowchart for demonstrating the abnormality determination processing performed by the data calculating part 160 of FIG. It is a figure which shows the result of having carried out FFT of the vibration measurement data obtained from the normal bearing in the environment with a high noise level. It is a figure which shows the result of having carried out FFT of the vibration measurement data obtained from a damage bearing in the environment with a high noise level. It is a figure which shows the result of having carried out the FFT of the vibration measurement data obtained from the normal bearing in the environment with a low noise level. It is a figure which shows the result of having carried out FFT of the vibration measurement data obtained from a damage bearing in the environment with a low noise level.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. In the following drawings, the same or corresponding parts are denoted by the same reference numerals, and the description thereof will not be repeated.

[Basic configuration of status monitoring device]
FIG. 1 is a block diagram showing the configuration of the state monitoring device according to the present embodiment. Referring to FIG. 1, the state monitoring apparatus 100 receives a signal from a vibration system sensor 20 installed in the device under test 10, monitors the state of the device under test 10, and detects an abnormality. The device under test 10 is a facility including a rotating device installed in, for example, a factory or a power plant. The rotating device comprises a rolling bearing 12. The rolling bearing 12 includes an inner ring 16 fitted to the rotary shaft 19, an outer ring 14 fixed to the device under test 10, and a plurality of rolling elements 18 disposed between the inner ring and the outer ring.

  The vibration system sensor 20 can detect abnormal vibration generated at the time of rotation. In the present embodiment, although the acceleration is exemplified as the monitoring target, the vibration system sensor 20 may be an acoustic sensor, an ultrasonic sensor, or an AE sensor other than the acceleration sensor.

  The state monitoring device 100 includes an amplifier 110, a filter 120, an A / D converter 130, a data acquisition unit 140, a storage device 150, a data operation unit 160, and a display unit 170.

  The voltage waveform (hereinafter referred to as an oscillating voltage waveform) of the vibration sensor 20 installed in the rotating machine is amplified by the amplifier 110, and the filter 120 removes band signals unnecessary for analysis and passes only necessary bands. (Passband 1 kHz to 10 kHz) and envelope processing are performed. The A / D converter 130 receives the output signal of the amplifier 110. The data acquisition unit 140 receives the digital signal from the A / D converter 130 and records measurement data in the storage device 150. The data operation unit 160 reads out the measurement data measured from the storage device 150, performs FFT analysis, and performs each FFT (to be described later) on the frequency spectrum of the low frequency band. Calculate the Σm value and the combined power ratio mm / Σo).

  The data calculation unit 160 determines the presence or absence of an abnormality of the device under test 10 from each feature amount. When the data calculation unit 160 determines the presence or absence of an abnormality, the data calculation unit 160 causes the display unit 170 to display the determination result.

[Description of abnormality determination processing]
FIG. 2 is a flowchart for explaining the abnormality determination process performed by the data calculation unit 160 of FIG. Referring to FIG. 2, first, in step S 1, data operation unit 160 reads from storage device 150 the vibration data, the rotational speed signal stored together with the vibration data, and the design specification of the monitoring target bearing.

  Subsequently, in step S2, the data calculation unit 160 calculates the fundamental frequency and the harmonic frequency of the vibration caused by the damage of the bearing based on the rotation speed signal and the design specifications.

  In the case of bearing damage, any one of the outer ring, the inner ring, and the rolling elements is damaged. For example, when the outer ring is damaged (groove defect), impulse noise is generated when the rolling element passes the damage. This impulse noise periodically appears as a function of rotational speed.

For example, the frequency of vibration when there is a defect in the outer ring is fout, the rotational frequency of the shaft is fr, the number of rolling elements is Z, the contact angle is α, the diameter of the pitch circle is D, and the diameter of the rolling elements is d. , Fout can be calculated by the following equation.
fout = 1/2 × Z × fr × (1−d / D × cos α)
However, since the impulse noise is different from the sine wave, the vibration waveform includes harmonic components. Therefore, if FFT is performed, harmonics of integer multiples thereof can be observed with fout as the fundamental frequency. When the inner ring has a defect, the rolling element has a defect, and the cage has a defect, the fundamental frequency can be calculated by a known equation as in the case of the outer ring.

  Subsequently, in step S3, the data calculation unit 160 determines the main monitoring band and the sub monitoring band. The main monitoring band is a band defined around the above-mentioned fundamental frequency and the frequency of harmonics, and the sub monitoring band is a band other than that. That is, the data calculation unit 160 sets the damage frequency and the frequency component as a target by taking the frequency component and its higher order components as a target due to the damage of the rotating part calculated based on the rotation speed signal and the design specifications. A main monitoring band of ± several% (about ± 5% to ± 15%) of the damage frequency fundamental wave is set in the higher order component.

  For example, it is preferable to set the monitoring band such that 30% of the monitoring band occupies 30% of the main monitoring band and 70% of the sub monitoring band, and 10% of the monitoring band occupies the main monitoring band, It is more preferable to set the monitoring band so that 90% occupies the monitoring band. In this case, the ± 5% band of the fundamental frequency around the fundamental frequency and the harmonic frequency is the main monitoring band, and the remaining part is the sub-monitoring band.

  Subsequently, in step S4, the data calculation unit 160 executes FFT on the acquired vibration measurement data, calculates Σm and Σo in step S5, and calculates combined power ratio mm / Σo in step S6.

  In the above, the combined power of the frequency components of the measured data in the main monitoring band is 実 測 monitoring (Σm). Σ m means the product sum of frequency component power spectra of the main monitoring band.

  Also, let otherothers (Σo) be a combined power of frequency components of measured data outside the main monitoring band. Σ o means the product sum of frequency component power spectra of the sub monitoring band.

  By utilizing such proposed analysis parameters (Σm, oo, and combined power ratio mm / Σo), discrimination of bearing abnormalities is facilitated even for vibration measurement data into which low frequency noise is mixed, While the reliability of the abnormality diagnosis becomes high, it becomes possible to make an abnormality diagnosis automatically.

  In step S7, data operation unit 160 determines whether combined power ratio mm / Σo is larger than a threshold. In step S7, when the Σm / 成立 o> threshold value is established (YES in S7), the data calculation unit 160 determines that the bearing is abnormal in step S8. On the other hand, if Σm / Σo> threshold value is not established in step S7 (NO in S7), data operation unit 160 determines in step S9 that the bearing is normal.

  If one of the determinations is performed in step S8 or step S9, the abnormality determination process ends in step S10.

  The abnormality diagnosis method according to the present embodiment described above determines the presence / absence of abnormality of the bearing based on measurement data obtained by measuring mechanical vibration generated at the time of rotation of the bearing by the vibration system sensor 20. This abnormality diagnosis method comprises the steps of setting the main monitoring band to each of the fundamental frequency and the harmonic frequency of the fundamental frequency caused by the damage of the bearing calculated based on the rotational speed signal and the bearing design specifications (S3), and measurement A first value に 対 し て m indicating the sum of the power of frequency components in the main monitoring band at each of the fundamental frequency and the harmonic frequency by performing frequency analysis on the data and the frequency in the sub monitoring band other than the main monitoring band Determining the abnormality of the bearing to be diagnosed using a second value oo indicating the sum of component powers (S5 to S9).

  Preferably, the determining (S7 to S9) determines that the bearing to be diagnosed is abnormal if a value mm / Σo obtained by dividing the first value mm by the second value oo exceeds a threshold value. .

  In addition, in combination with the conventional judgment, if it exceeds the threshold value of the OA value set in advance from the analysis result (OA value) of the vibration measurement data, it is judged to be in the "needs attention state" and the special introduced this time Detailed analysis and abnormality diagnosis may be performed by using various analysis parameters (Σm, Σo, and combined power ratio mm / Σo).

[Example]
Hereinafter, since the determination is performed under the four conditions by applying the abnormality determination method described in the present embodiment, data will be shown and described.

Test conditions and analysis methods are shown below in a) to d).
a) Bearing measurement at low speed rotation (100 rpm) vibration measurement b) Bearing (normal bearing, damaged bearing: outer ring raceway surface damaged at one point)
c) Environment (high noise level / low noise level)
d) Analysis method: For vibration data, after BPF (1 kHz to 10 kHz) processing, envelope processing and FFT analysis were performed. After the analysis, the OA value, the mm value, and the combined power ratio mm / oo were calculated for the frequency spectrum of the low frequency band (〜200 Hz).

  FIG. 3 is a diagram showing the result of FFT of vibration measurement data obtained from a normal bearing under an environment where noise level is high. In FIG. 3, a main monitoring band M1 corresponding to a fundamental frequency and main monitoring bands M2 to M14 corresponding to harmonic frequencies are set. The proportion of the main monitoring band occupying the entire monitoring band was set to 10%, and the proportion of the secondary monitoring band to 90%. In this case, the calculation results of the respective numerical values are OA = 0.0968, Σm = 0.0456, and Σm / Σo = 0.53.

  FIG. 4 is a diagram showing the result of FFT of vibration measurement data obtained from a damaged bearing under an environment of high noise level. The main monitoring band and the secondary monitoring band were set as in FIG. The calculation results of the respective numerical values were OA = 0.1728, Σm = 0.1233, and Σm / Σo = 1.01.

  FIG. 5 is a diagram showing the result of FFT of vibration measurement data obtained from a normal bearing under a low noise level environment. The main monitoring band and the secondary monitoring band were set as in FIG. The calculation results of the respective numerical values were OA = 0.0223, Σm = 0.0106, and Σm / Σo = 0.54.

  FIG. 6 is a diagram showing the result of FFT of vibration measurement data obtained from a damaged bearing under a low noise level environment. The main monitoring band and the secondary monitoring band were set as in FIG. The calculation results of the respective numerical values were OA = 0.0694, Σm = 0.0555, and Σm / Σo = 1.33.

  The OA values used in the conventional analysis method are the data obtained from normal bearings in a high noise level environment (FIG. 3) and the data obtained from damaged bearings in a low noise level environment (FIG. 6) Higher than). Therefore, it is understood that automatic determination of a damaged bearing can not be made by threshold setting for the OA value.

  Here, when reanalysis is performed using analysis parameters (Σm, 合成 o, and combined power ratio mm / Σo), mm value and combined power ratio mm / Σo are normal when the noise level is low and the noise level is high in the damaged bearing. Appears at higher values than bearings. In particular, with regard to the combined power ratio mm / 分析 o, the analysis results of the normal bearings shown in FIGS. 3 and 5 are substantially the same level (0.53 and 0.54) regardless of the noise level. There is a remarkable difference from the analysis results (1.01, 1.33) of the damaged bearing shown in For example, when the threshold value is set to about 0.7 to 0.8 for the combined power ratio mm / mo, it is possible to automatically distinguish between a damaged bearing and a normal bearing. Therefore, if the threshold value is appropriately determined for the combined power ratio mm / Σo, it is possible to accurately determine the damaged bearing and the normal bearing.

  Summarizing the above results, when there is noise contamination in the damage frequency band, comparing the OA value, the OA value of the damaged bearing measured in a low noise environment is the OA of a normal bearing measured in a noisy environment Since it is lower than the value, automatic judgment can not be made by setting the threshold. Therefore, automatic determination has a high possibility of erroneous determination and is not practical.

  On the other hand, when the analysis parameters proposed in the present embodiment were introduced, it was found that the mm value of the damaged bearing was higher than the mm value of the normal bearing when the mm values were compared. Further, when the Σm / Σo values are compared, there is a clear difference in the combined power ratio of the damaged bearings from 1.01 to 1.33 with respect to the combined power ratio of the normal bearings from 0.53 to 0.54. Therefore, according to the determination method of the present embodiment, it is possible to automatically determine the vibration data of the bearing measured in an environment where the noise level is high.

  It should be understood that the embodiments disclosed herein are illustrative and non-restrictive in every respect. The scope of the present invention is shown not by the above description of the embodiments but by the scope of claims, and is intended to include all modifications within the meaning and scope equivalent to the scope of claims.

  10 Device under test, 12 bearings, 14 outer rings, 16 inner rings, 18 rolling elements, 19 rotating shafts, 20 vibration system sensors, 100 condition monitoring devices, 110 amplifiers, 120 filters, 130 converters, 140 data acquisition units, 150 storage devices, 160 Data operation unit, 170 display unit, M1 to M14 Main monitoring band.

Claims (5)

An abnormality diagnosis method of determining the presence or absence of an abnormality of the bearing based on measurement data obtained by measuring mechanical vibration generated at the time of rotation of the bearing by a vibration system sensor,
Setting a main monitoring band for each of the fundamental frequency attributable to the damage to the bearing calculated based on the rotational speed signal and the bearing design specifications and the harmonic frequency of the fundamental frequency;
A first value indicating a sum of powers of frequency components in the main monitoring band at each of the fundamental frequency and the harmonic frequency by performing frequency analysis on the measurement data; Determining an abnormality of a bearing to be diagnosed using a second value indicating a sum of powers of frequency components in the monitoring band.   The abnormality diagnosis according to claim 1, wherein the determining step determines that the bearing to be diagnosed is abnormal when a value obtained by dividing the first value by the second value exceeds a threshold. Method.   The abnormality diagnosis method according to claim 1, wherein the vibration system sensor is at least one of an acceleration sensor, an acoustic sensor, an ultrasonic sensor, and an AE sensor.   The state monitoring apparatus which diagnoses abnormality using the abnormality diagnostic method in any one of Claims 1-3.   The condition monitoring device according to claim 4, further comprising a filter unit that removes a signal of a frequency band not to be analyzed from the signal waveform detected by the vibration sensor.

Images